The Shewhart-Deming Cycle

The Shewhart-Deming

CycleThe key is a continuous cycle of improvement

Act

Plan

Do

Study

Definition Six Sigma is a structured

strategic quality management approach that: enables improved decision

making by the leadership using tools based on sound

business, statistical, and engineering principles.

Evolution of Six Sigma Originally from Motorola company focused

on manufacturing processes – mid 1980s. Six Sigma has evolved to become

applicable to non-manufacturing processes as well:

Services Medical and Insurance Procedures Call Centers Human Resources

GoalThe primary goals of Six Sigma: increase customer

satisfaction business profitability through reduction/elimination of

defects.

Known Industrial Applications Leaders in the successful

implementation of Six Sigma Motorola, GE, Allied Signal, Microsoft, Intel,

3M, Boeing, etc. WIPRO, GE India, Philips India, ONGC,

Adidas, Baxter India Manufacturing (quality improvement,

cost reduction, process optimization, measurement error improvement, safety improvement, new product development, etc.)

Known Industrial Applications Transactional: Improving customer

OTIF (On-time, In-full), call centers, supply chain excellence, human resource management, library management, payroll accuracy, etc.)

Business: Resource allocation, optimizing and prioritizing capital investments, new business development/ new product introduction, etc.

Functional Areas of Application

Improve ProductsImprove Products Design to process

capability Design to customer

needs Improve MaterialsImprove Materials

Reduce variability Improve delivery

time Reduce cost

Improve PeopleImprove People Skills and capability Problem solving

methods Improve ProcessesImprove Processes

Remove variability Improve stability Match process to

customer needs

Concepts

The Inspection Exercise

Count the number of times the 6th letter of the alphabet appears in the following text:

The necessity of training farm hands for the first class farms in the fatherly handling of farm live stock is foremost in the eyes of the farm owners. Since the forefathers of the farm owners trained the farm hands for first class farms in the fatherly handling of farm live stock, the farm owners felt they should carry on with the family tradition of training farm hands of the first class farmers in the fatherly handling of farm live stock because they believe it is the basis of good fundamental farm management.

Count the number of times the 6th letter of the alphabet appears in the following text:

The necessity in training hired hands in the strange handling of valuable live stock in premier operations is a priority in the eyes of the operations owners. Since the ancestors of the owners trained the hired hands in premier operations in the strange handling of valuable live stock, the operations owners thought they should carry on with the happy tradition of training hired hands in the premier operations in the strange handling of valuable live stock because they believe it is the basis of good basic operations management.

The Inspection Exercise

Reality of Escaping Defects

TToottaall DDeeffeeccttss//UUnniitt

EE sscc aa

pp ii nngg D

D eeff ee

cc tt ss

Escaping Defects

It doesn’t matter how good your inspection and test processes are; the more defects you create, the more defects escape to the customer

What does “Six Sigma” mean?

Six Sigma stands for Six Standard Deviations from the mean. Six Sigma allows for only 3.4 defects per million opportunities for each product or service transaction. For a process that is centered on the

target, the lower and upper specification limits are each at a distance of 6 standard deviations from the mean.

Sigma vs. Defect Levels

2 308,5373 66,8074 6,2105 2336 3.4

PPM

Sigma Level

Defects perMillion

Opportunities

The Six Sigma Measurement

What Does 6 Sigma Mean In Your Daily Life ?

Tax Advice

7

Sigma Level

1,000,000

100,000

10,000

1,000

100

10

1

PPM

Restaurant BillsPayroll Processing

Prescription Writing

Baggage Handling

AirlineSafety Rate

3 4 5 621 3 4 5 621

Practical Meaning

99% Good 99.99966% Good

Postal System

20,000 Lost Articles of Mail / Hr 7 Lost Articles / Hr

Airline System

Two Short-Long Landings / Day 1 Short-Long every 5 Years

Medical Profession

200,000 Wrong Drug Prescriptions / Year 68 Wrong Drug Prescriptions / Year

Key Benefits & Areas of Application

Data-driven decision making Quality Improvement Waste reduction Reduction of factory costs Increase in productivity and capacity Product and process optimization Alignment between “voice of customer”

and “voice of business/process” Meet business and customer schedules Supply Chain Excellence

A Statistical View Everything is a process

(manufacturing, financial operations, call centers, medical procedures, insurance procedures, etc.).

Every process has inherent variation.

A Statistical View (continued) All efforts to understand

variation and improve the process must be data driven. It is not sufficient to make

improvements based on average only; due attention must be paid to variation or spread

Two Types of Six Sigma: DMAIC

Define the opportunities (create the project charter, identify savings opportunity, project metrics, etc.)

Measure (current/baseline) performance Analyze opportunities (develop more

detailed product & process understanding)

Improve performance Control performance (“maintain the

gains”)

Two Types of Six Sigma: DMAIC

DMAIC Primarily used for existing processes and

products (not new product development) Reactive in nature Involves achieving higher quality, yield, etc.

Systematic application of statistical thinking based on trustworthy data

(Similar to PDCA approach “Plan, Do, Check, Act”)

Two Types of Six Sigma: DFSS

DFSS (Design For Six Sigma) is aimed at building

a high level of quality into the design of the product at the very beginning.

Extensive data collection in terms of customer needs, translating the knowledge into manufacturing parameters, and conducting upfront reliability and quality testing on prototypes to assure delivery on quality and cost requirements.

The Funneling Effect

Optimized Process

> 30 Inputs

8 - 10

4 - 8

3 - 6

Found Critical X’s

Controlling Critical X’s

10 - 15

All X’s

1st “Hit List”

Screened List

MMEASUREEASURE

AANALYZENALYZE

IIMPROVEMPROVE

CCONTROLONTROL

DDEFINEEFINE

Define Phase - PurposePurposeTo develop a fully-defined projectSome key steps: Drive project selection through

business strategy and the business objectives

Carefully analyze and incorporate customers needs

Define the project objectives Scope the project appropriately

Measurement Phase - PurposePurpose

To define the current process and establish metrics

Some key steps: Process mapProcess map Process outputs and inputs Process outputs and inputs (Y’sY’s and X’sX’s) Cause and Effects MatrixCause and Effects Matrix How “trustworthy” are our measurements?How “trustworthy” are our measurements? How capable is our process in meeting How capable is our process in meeting

customer requirements?customer requirements?

Analyze Phase - PurposePurpose Learn about the relationships

between the X’s and Y’s Identify potential sources of

process variabilitySome key steps: Failure Mode and Effects AnalysisFailure Mode and Effects Analysis Multi-Vari StudiesMulti-Vari Studies Plan the first improvement

activities

Improvement Phase - PurposePurpose

• Quantify relationship between critical X’s and Y’s

Some key steps: Experimentation and Data Collection Design of ExperimentsDesign of Experiments (DOE) is the

backbone of this phase

Control Phase - PurposePurpose• Maintain the gains!!!Some key steps: Document and implement a control control

planplan Establish and monitor long-term long-term

capabilitycapability Implement continuous improvementcontinuous improvement

efforts

Tools

8 Key Tools Process MapsProcess Maps: to document key process

steps, inputs, and outputs Cause and Effects MatrixCause and Effects Matrix: to prioritize key

inputs for action Measurement Systems AnalysisMeasurement Systems Analysis: to evaluate

accuracy and precision of measurement systems

Capability StudiesCapability Studies: to establish how process performance compares to customer specifications

8 Key Tools (continued) Failure Modes and Effects Analysis (FMEA)Failure Modes and Effects Analysis (FMEA): to

identify high risk inputs and improvement actions

Multi-Vari StudiesMulti-Vari Studies: to provide quantitative clues for identifying inputs to leverage

Design of Experiments (DOE)Design of Experiments (DOE): to systematically study process inputs to identify optimal process windows

SPC and Control PlansSPC and Control Plans: to monitor the process and document all actions necessary to maintain world class performance

Process Map

C&E (Cause & Effect) Matrix

Measurement Systems Analysis (MSA)

Data Integrity Validity: Is the “right” aspect of the process being measured?

The data can be from a very reliable method or source, but still not match the operational definitions established for your project

Measurement Systems Analysis (MSA)

Data Integrity Reliability: deals with the accuracy

and consistency of the data For data processing and handling

situations, use audits to assess reliability When the data comes from test

equipment, use a Gage R&R study When the data are from a qualitative

assessment, use an Attribute MSA study

Measurement Systems Analysis (MSA)

Measurement System: Is the measurement system producing good data? Resolution: Smallest unit of measurement

displayed by the system/instrument. Accuracy: Does the average of the

measurements deviate from the true value? Bias: The difference between the average of

all repeated measurements that might be made on a sample at a given time, and its true value.

Measurement Systems Analysis (MSA)

Repeatability: Variation that occurs when repeated measurementsrepeated measurements are made of the same variable under absolutely identical conditions.

Reproducibility: The difference in the average of the measurements made by:

Different people Same instrument Measuring the same characteristic Different conditions (environment, time, etc.)

Process Capability Indices

Capability Ratio - compares the capability of a process (voice of the process) to the specification limits (voice of the customer).

USLUSLLSLLSL

Voice of the Customer

Voice of The Process

USLUSLLSLLSL

Voice of the Customer

Voice of The Process

Voice of the CustomerVoice of the Process

Voice of the CustomerVoice of the Process

Process Capability Ratios, Cp

SpreadProcess Tolerance TotalCP

6sLSL-USLCP

Process of VoiceCustomer of VoiceCapability

Two Groups of Capability Indices

C - Represents process capability - what the process potential is given a stable process Standard deviation estimated from Moving Range or

pooled standard deviation – represents common cause variation

P - Represents process performance - what has happened, not necessarily what will happen Standard deviation estimated from the traditional

formula – includes both common and special causes of variation

Illustration of CpCp = 0.5

Cp = 1.0

Cp = 1.5

Cp = 2.0

Process Capability Indices - Cp, Pp

Sigma6pec - Lower SUpper Spec

Cp

Sigma is estimated from a control chart (common cause variation)

PpSigma is estimated as the traditional standard deviation (common and special cause variation)

Stable Process

Unstable Process

2dMRσ ˆ

1n)X(X

σ2

i

ˆ

Cp & Cpk for an Off-Center Process

Cp= 1.3

Cpk = 1.3

Cp= 1.3

Cpk = 0.8

Cp= 1.3

Cpk = 0.0

FMEA - Definition Primary objectivePrimary objective::

Identify ways in which the process input variables (Xs) can fail and determine what effect that has on process outputs (Ys)

A structured approach:A structured approach: Estimates the risk associated with specific causes

How X fails and the impact on Y Prioritizes the actions that should be taken to

reduce the risk How to prevent X from failing

Potential inputs into the control plan

History of FMEA Originally used in the 1960s for the

Apollo space missions.

In 1974 the US Navy developed FMEA standards.

In the late 1970s, automotive industry adapted FMEAs mainly driven by liability costs.

Types of FMEA SystemSystem - used to analyze systems and sub-

systems in the early concept and design stages

Focuses on potential failure modes associated with the functions of a system caused by the design

DesignDesign - used to analyze product designs before they are released to production

Focuses on product function ProcessProcess - used to analyze potential failures in

existing processes Focuses on process inputs

OverviewProcess

Step/Input Potential Failure Mode Potential Failure EffectsSEV

Potential CausesOCC

Current ControlsDET

RPN

Actions Recommended

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

What is the Input

What What can go can go wrong wrong

with the with the Input?Input?

What can What can be done?be done?

What is What is the Effect the Effect

on the on the Outputs?Outputs?

What are What are the the

Causes?Causes?

How are How are these found these found

or or prevented?prevented?

How How Bad?Bad?

How How Often?Often?

How How well?well?

Ways of Learning Passively Passively - Observe naturally

occurring informative events (Multi-vari Studies) If you’re lucky, an informative event

might happen while you’re watching ExperimentallyExperimentally - Create

informative events

Ways of Learning (continued)

Experimental DesignExperimental Design Proactively manipulate input

variables to study the effect on the output variables

Invites informative events to occur Experiments, if done correctly, are

Efficient Powerful

Multi-Vari Studies - Purpose

Collect data to understand which Xs affect Ys Y = f(X)

Collect data to support intuition that created C&E and FMEA

May discover inputs initially omitted from process map or removed during C&E / FMEA

Find the key Xs to advance to the Improve phase and avoid… Wasting time experimenting with unimportant Xs Implementing control plans on unimportant Xs

Multivari Analysis Before initiating DOEs, Multivari analysis is

designed to extract the maximum amount of information using either planned data collection or using existing historical data.

Concepts like ANOVA (Analysis of Variance) and other statistical methods used to identify and quantify relationships between key X’s and Y’s.

Less expensive than DOEs but this may often lead to identification of correlation rather than establish cause and effect.

Design of Experiments A systematic method for using data to

understand the cause and effect relationships in a process

Study the effects of simultaneously changing more than one X

Changes are deliberately made to input variables (Xs) to observe changes to the output response(s) (Ys)

Design of Experiments (continued)

Seek influential Xs which Center the Y on the target Minimize the variability of Y Minimize the effect of Noise Variables

Advantages of Designed Experiments Enable data-based decisions Create understanding of the process

and how to control it Take into account the inherent noise in

the system Efficient - maximum information with

minimal effort Detect variable interactions

SPC – Statistical Process Control

Used to: Quantify the “normal” variation in the process

parameter when process is in a state of control

Identify the major sources of variation Identify the correct type of control chart Identify process owners responsible for

regular monitoring and maintenance, updates to the chart, etc.

Establish clear reaction plans based on signals/alarms generated by the control chart

Control Plan How to maintain the gains? Often if the process is left to itself after

the initial optimization effort, it may revert to poor performance over time.

An effective control plan identifies monitoring schemes, owners accountable for fluctuations in process performance, and a process in place to repeat the DMAIC process if necessary.

Applications

Known Industrial Applications Leaders in the successful

implementation of Six Sigma Motorola, GE, Allied Signal, Microsoft, Intel,

3M, Boeing, etc. WIPRO, GE India, Philips India, ONGC,

Adidas, Baxter India Manufacturing (quality improvement,

cost reduction, process optimization, measurement error improvement, safety improvement, new product development, etc.)

Known Industrial Applications Transactional: Improving customer

OTIF (On-time, In-full), call centers, supply chain excellence, human resource management, library management, payroll accuracy, etc.)

Business: Resource allocation, optimizing and prioritizing capital investments, new business development/ new product introduction, etc.

Keys to Six Sigma Success Identify critical business issuescritical business issues and

strategies LinkLink efforts (projects) to resolution

of roadblocks Set clear expectationsexpectations for results MeasureMeasure the progress (Metrics) Manage for resultsresults